Inclusive Jet Cross Sections in Photoproduction

The measurement of jet production in photoproduction (gamma + p -> jets + X) allows tests of the standard theory of strong interactions, Quantum Chromodynamics (QCD) and of our understanding of the proton and the photon in terms of the quarks and gluons from which they are constructed. The idea that the photon has quark and gluon structure sounds strange at first. It arises because quantum field theory allows the photon to split into a quark-antiquark pair for a brief period of time. The resulting quarks can also radiate gluons, which in turn can split into more quarks and antiquarks, eventually resulting in a rather complex object, very different from the simple point-like photon from which the process started. If the photon collides with a proton during one of these fluctuations, the quark and gluon structure can be sampled through the interactions that take place.

This paper is concerned with the quarks and gluons that are produced in the collisions of photons with protons. It is not possible to observe these outgoing particles directly, since they quickly `hadronise' into jets of particles. However, since the properties of the jets are closely correlated with those of the quarks and gluons, it is enough to study the jets. The paper follows another recent analysis of jet photoproduction, in which pairs of jets was studied. However, in the present paper, all events in which jets are reconstructed are considered, even if there is only one. Although it implies that the kinematics of the events are less well constrained, requiring only one jet has the advantages of boosting the statistics, allowing access to an extended range in the kinematic variables and naturally avoiding regions that are difficult for theoretical calculations. Compared to the last H1 measurement of inclusive jet production, the statistics are improved by nearly 2 orders of magnitude!

Fig 4 of paper The cross sections for jet production are measured as a function of the jet pseudorapidity (essentially its angle in the detector relative to the beam axis) and its energy projection transverse to the beam axis. In order to test the theory as closely as possible, they are also measured multi-differentially in these variables and the photon-proton centre-of-mass energy. An example measurement of the transverse energy dependence is shown in the figure. Very high transverse energies of nearly 70 GeV are reached and the cross section drops by 6 orders of magnitude over the range of measurement. As can be seen from the bottom part of the figure, the state-of-the-art theory (`NLO' QCD) does a very good job of reproducing this variation using appropriate sets of quark and gluon densities for the proton and the photon. This is not the case when the slightly less refined version of the theory (`LO' QCD) is used. This good description of the data by NLO QCD is observed for all distributions studied, with the possible exception of the pseudorapidity distribution at the lowest transverse energy. Comparing the size of the errors on the data points with the bands representing the errors on the calculations, it is clear that theoretical improvements as well as better data are required in order to test the theory in even finer detail.

The data in the paper are also compared with similar measurments from proton-antiproton collisions in order to study the effects arising from the different structure of the photon and proton. This is done as a function of a variable x_T, which is related to the fractions of the photon and proton momenta carried by the interacting quarks and gluons. At low values of x_T, the results from photon-proton and from antiproton-proton collisions show a remarkably similar shape, illustrating that the photon really does develop into something similar to a hadron in its quantum fluctuations. At larger x_T, where the photon acts in a more point-like way, the distributions become very different.